This project was part of my capstone course for Computer Science. The project was built using Python and several data science libraries to create data visualizations, displaying traffic accident severity in Denver.
-
To launch the project, click the Binder button
, or go to https://bit.ly/3dGPqZZ.
-
Once the page is loaded, click the Voila button to render the project.
(If Voila isn't working in your browser, go to the Menu and click: Kernel -->> Restart & Rerun All)
Problem Summary
The City of Denver is currently facing challenges in accurately and efficiently assigning resources to different parts of the city. The City must soon decide where to allocate law enforcement and emergency services professionals in order to sufficiently cover each area of the city in proportion to the amount of crime or emergency scenarios seen in each area. With all the traffic accident data that the City of Denver has already collected over the years and continues to collect, it is possible to build a dashboard to assist the City in determining how to best allocate resources towards problematic areas. This dashboard, with the use of the existing traffic accident data, will display summarized information that the City of Denver can use to make further decisions regarding its resource allocation.
Application Benefits
The application’s major benefit includes the ability to summarize very large amounts of traffic accident data into a digestible and actionable format for the City of Denver to use in decision making. The application will be capable of ingesting large amounts of data and performing calculations based on appropriate algorithms. The application will be capable of outputting easy-to-understand information and visuals that allow users to understand any major factors that contribute to traffic accidents.
The data that drives this analytics dashboard can also provide crucial insights into future decisions in resource allocation and hiring. By analyzing the data from multiple angles, the dashboard will provide the ability for City employees to make informed business decisions on how to best place its resources and understand whether there is sufficient coverage for problematic areas.
Application Description
The proposed solution will be comprised of a backend service built with Python that utilizes data science libraries such as pandas, numpy, and sklearn to create data visualizations and output statistically significant factors. The analytics dashboard interface will also use the plotly library to display charts. In addition, the front-end user interface will be created with a Jupyter Notebook to enable easy and quick sharing or modification of data inputs.